[A89] Re: Clarification
[A89] Re: Clarification
TI did indeed put new limitations in on HW2 calcs. My HW2 89
reports an "Invalid relocation data in ASM program" error using
an (older) version of TTStart. 2.08 was indeed recalled due to
timing problems with the black G/L, which I have. I notified
TIemail@example.com about this shortly before they recalled it.
They told me that they did not recall it because of my
complaints (regarding "features" and possible improvements),
but because of the timing errors.
I think you're correct about that extra 64k. My total amount
of free flash decreased after I installed it.
IT will probably release somthing like AMS 2.12, etc.
However, you may be wrong about the time. I have noticed that
if you reset often (like once every five days or more), the
clock will be of by several minutes.
When's Command Post coming out? It's already December. ;)
-Duncan Smith (slimey_limey)
mrslimeylimey AT hotmail DOT com
>From: "Greg Dietsche" <Gregory.Dietsche@cuw.edu>
>Subject: [A89] Clarification
>Date: Fri, 27 Dec 2002 15:01:55 -0600
> to clear a few things up :)
>1. KerNO (of which I'm the author) does not support kernel based programs,
>and never will. So, I prefer to call it a "kernel like" program or more
>simply stated: The Kernel that Never Was, Never Will Be, and Always Shall
>Be. If you are looking for a good kernel to run kernel based programs, i
>recommend PreOS, or UniversalOS. KerNO is designed to work with NOSTUB
>2. AMS 2.08 was recalled because of problems with linking and the black
>cable AFAIK. Hopefully, TI will be re-releasing it soon :) I have 89, 92p,
>and v200 copies from the first time they released it.
>3. The clock feature on AMS 2.08 is run from Auto Interrupt 3, and is
>extremely accurate on HW2 calcs (based off a RTC crystal). HW1 calcs do not
>have this, and the AI3 frequency fluctuates around 0.616322/second versus
>1/second on my TI-89 HW1. ie, in a 60 second time frame, my hw1 calc will
>trigger ai3 ~87 times.
>IF an older asm game is run that uses OSSetSR(0x700) to disable interrupts
>for row reading during game play, the clock will loose time. In this case,
>the clock will loose time because AI3 was not allowed to execute its 3 or
>so lines of ASM and update the system clock by adding one second to the
>time. So don't be surprised if you run an older game, and it makes the
>clock loose time.
>In my mind, the best way to _rowread now, is to disable the necessary auto
>interrupts (1 and 5) by redirecting them to a dummy handler (rte) rather
>than using ossetsr.
>4. Large ASM games on 2.08 are not broken. If you are having problems with
>PPG compressed programs, it is because of a bug in the launcher, NOT
>because of new limitations. Solution: get the latest version of ttstart,
>and your programs will run.
>5. AMS 2.08 is very nice for the TI-92p/V200 HW2 calcs otherwise, don't
>bother. (I own a 89 hw1, 92p hw2, and a v200) Additionally, i believe that
>ams 2.08 takes an extra 64k of flash on the ti-89 (not on the 92p/v200) due
>to sector boundaries.
>GForce Programming: http://calc.gregd.org
>Detached Solutions: http://www.detachedsolutions.com
STOP MORE SPAM with the new MSN 8 and get 3 months FREE*.